Commit Graph

260 Commits

Author SHA1 Message Date
Scott LaVarnway
5fe0e55ca4 Merge "Eliminated frame_type check in get_partition_probs()" 2015-06-16 13:40:23 +00:00
Scott LaVarnway
b2658ec321 Eliminated frame_type check in get_partition_probs()
Moved the frame_type check to the tile level and stored
the prob ptr in MACROBLOCKD.

Change-Id: I10b5a4abd58213dc7610e3ade1a1583c01526842
2015-06-16 05:37:54 -07:00
Scott LaVarnway
a41fe749a8 Merge "Update use_prev_frame_mvs flag in decoder." 2015-06-16 12:28:46 +00:00
James Zern
15c839f563 decode_tiles_mt: remove incorrect TODO
all allocated workers are used, the final one in the main thread.

Change-Id: I04647d4fb2e01d6d404790e7899515289047f553
2015-06-10 15:56:51 -07:00
James Zern
b105414118 vp9_decodeframe: simplify init_read_bit_buffer signature
Change-Id: I342075eb5a4ba2a85d36d47ae52d1f3476039e31
2015-06-10 15:53:30 -07:00
James Zern
92146eb8ac vp9_decode_frame: remove explicit read_bit_buffer init
this is done by init_read_bit_buffer()

Change-Id: I222eaa4e9758ff9f7e1e4122106c5c4652ffa99c
2015-06-10 15:50:54 -07:00
James Zern
b0bafd0439 vp9_decodeframe: move public funcs to end of file
Change-Id: Iaea9850890b726c7b5552c5f02b3a309086edc85
2015-06-10 15:47:57 -07:00
James Zern
38dd0448cd vp9_decodeframe: reorder some functions
removes the need for a prototype for the static function
dec_build_inter_predictors_sb

Change-Id: I36580ceae061d27f341ab0a16ece479f92e98004
2015-06-10 15:44:21 -07:00
James Zern
587bd3669e vp9_decodeframe: hide vp9_dec_build_inter_predictors_sb
+ strip 'vp9_' prefix

Change-Id: Ia057432095e5741473275d4da03ab665c37e924e
2015-06-10 15:31:55 -07:00
Scott LaVarnway
c9976b32b4 Update use_prev_frame_mvs flag in decoder.
Added check to see if last frame was all intra.  This will
eliminate two checks in find_mv_refs_idx().  Also, do not
update the frame mvs if the current frame is all intra.

This improved performance on material with frequent
intra-only frames.

Change-Id: I44a4042c3670ab0d38439d565062a0e2a1ba9d1e
2015-06-08 03:38:13 -07:00
hkuang
87c21a95ae Optimize the decode_partition. About 0.7% gain on N10.
Change-Id: Ia689c254bd2d4f274abcc451a9b758f62e3a2b1f
2015-06-05 15:47:09 -07:00
Frank Galligan
bfb6d48812 Add control to skip loop filter in VP9 decoder.
This control allows the application to skip the loop filter in the
decoder. This is an advanced control that should only be used in
extreme circumstances as it may introduce and accumulate decode
artifacts.

Change-Id: I278c65c60826f84c9141ebe06c6eeed3c2335fa8
2015-06-05 10:07:09 -07:00
Scott LaVarnway
8bb37dd069 Remove cm parameter from vp9_decode_block_tokens() part 2
Change-Id: Iee24b6bb095f748333223e6036fc5c9d9e7e5f1c
2015-06-04 07:13:19 -07:00
Scott LaVarnway
f779dba405 Remove counts param
Moved to MACROBLOCKD.

Change-Id: Icce765b334f2755f4fe2a4c39fb2ae2d7660d004
2015-06-02 09:06:00 -07:00
Johann
1e4473b216 Check size restrictions before running test vector
Change-Id: I60ea7724e6ab06fc658f678c1b76d984a43f5a5e
2015-05-27 15:55:17 -07:00
James Zern
85076fc5ab dec_build_inter_predictors: don't return a void fn
split call of extend_and_predict() and return, fixes visual studio build
warning since:
0a80164 Move mc_buf to cut down size of MACROBLOCKD.

Change-Id: I7cdf712941ef773a07f038539cb8080dc27861cd
2015-05-16 10:33:57 -07:00
James Zern
f3bf5f2029 vp9_decodeframe.c: make a function static
silences a missing declaration warning

Change-Id: I2f49ebca9ba7a47f3c48f5fe919b90cd4114a9bc
2015-05-15 10:43:47 -07:00
Frank Galligan
d610ead258 Merge "Move mc_buf to cut down size of MACROBLOCKD." 2015-05-15 15:20:39 +00:00
Frank Galligan
0a80164c94 Move mc_buf to cut down size of MACROBLOCKD.
Change-Id: Icea64b9e5632b41aaa7cd7018c501d6add9b7a7f
2015-05-14 19:10:02 -07:00
Johann
1d7ccd5325 Relocate memory operations for common code
With the sad functions, and hopefully the variance functions soon,
moving to the vpx_dsp location, place the defines used in the
reference C code in a common location.

Change-Id: I4c8ce7778eb38a0a3ee674d2f1c488eda01cfeca
2015-05-13 11:41:15 -07:00
Adrian Grange
5f0dc57653 Merge "Reduce border extension by 1 line" 2015-05-11 22:47:49 +00:00
Jingning Han
2b2b461d39 Sort variables dependency in read_uncompressed_header
Remove a few repeated data structure reads from
read_uncompressed_header.

Change-Id: I6eb741b39f9415ad0aa4631dfbf4a1ace4eba56a
2015-05-11 10:07:55 -07:00
Adrian Grange
0af5ff49bd Reduce border extension by 1 line
The code was using one row too many.

Change-Id: Ie0c05d02c22ae3d0d13d3b6565c40de3bc4fa17a
2015-05-06 14:44:24 -07:00
hkuang
92b199061a Correct the inter prediction coordinate calculation which greatly reduced the
times of border extension.

Change-Id: I8e5bd590cc696ee71cfe1f4cc66c12fb24aaf44e
2015-05-06 10:55:56 -07:00
James Zern
9e81112df2 vp9_decodeframe: simplify compare_tile_buffers
return the difference between the 2 buffer sizes rather than exactly
-1/0/1.

Change-Id: Idf1ccff7088b31845470bcc71bea5927b0598cc7
2015-04-29 17:42:30 -07:00
James Zern
f58011ada5 vpx_mem: remove vpx_memset
vestigial. replace instances with memset() which they already were being
defined to.

Change-Id: Ie030cfaaa3e890dd92cf1a995fcb1927ba175201
2015-04-28 20:00:59 -07:00
Scott LaVarnway
afcb62b414 WIP: Use LUT for y_dequant/uv_dequant
instead of calculating every block.

Change-Id: Ib19ff2546be8441f8755ae971ba2910f29412029
2015-04-28 07:52:06 -07:00
Scott LaVarnway
8b17f7f4eb Revert "Remove mi_grid_* structures."
(see I3a05cf1610679fed26e0b2eadd315a9ae91afdd6)

For the test clip used, the decoder performance improved by ~2%.
This is also an intermediate step towards adding back the
mode_info streams.

Change-Id: Idddc4a3f46e4180fbebddc156c4bbf177d5c2e0d
2015-04-21 11:16:45 -07:00
hkuang
7e8e507bfb Remove unnecessary mv clamp with on demand border extension.
Change-Id: Ia2956f06f409b9b0ca8320ca4c1ea5680e938402
2015-04-08 17:16:52 -07:00
Frank Galligan
5668dcc7b9 Refactor dec_build_inter_predictors
Refactor the loops in dec_build_inter_predictors to try and decrease
the number of instructions. Limited testing saw about 1% perf
increase on x86 and about 0.67 % perf increase on Arm.

Change-Id: I69cfe6335bb562fbaaebf43fb3f5c5a2a28882a2
2015-04-08 15:00:29 -07:00
hkuang
73c8fe5deb Fix error of "Left shift of negative value -1".
Change-Id: Ia4f3feb20df0e89cc51b02def858e12e927312cc
2015-04-02 17:35:33 -07:00
hkuang
915d5c12d3 Merge "Fix a tsan error bug in frame parallel decode." 2015-03-05 14:55:27 -08:00
hkuang
2f06f5087e Merge "Remove unnecessary loopfilter init." 2015-03-04 11:27:58 -08:00
Adrian Grange
3807dd82ab Make encoder buffer allocation dynamic
Frame buffers are now allocated dynamically on-demand.

Entries in the reference frame map, cm->ref_frame_map,
may now be set to -1 (INVALID_IDX) to indicate that
there is not a valid reference buffer in that "slot".

All slots in the reference frame map are now initialized
to the empty state (-1) and each buffer is initialized
to have a reference count of 0.

Change-Id: Id1afe98de98db4ae8b2dfefed7889c3b28c68582
2015-03-04 07:58:32 -08:00
hkuang
a723f5ecd1 Fix a tsan error bug in frame parallel decode.
A frame may be waiting for an out of border pixel from another
frame. A frame's row progress variable is set to -1 when start being decoded
and another frame may be waiting for -2 row pixel from this frame.
In this case, vp9_frameworker_wait will return directly and skip the waiting
which leads to tsan error between threads.

Change-Id: Id16604915fb598b823e34393f696e3aa46fb6422
2015-03-03 12:58:25 -08:00
hkuang
f9420b45e5 Remove unnecessary loopfilter init.
Loopfilters are init in vp9_decode_frame.

Change-Id: I4fbf6286b9b231451452e4ef3c19877a7b9a6768
2015-03-03 10:40:39 -08:00
Yaowu Xu
6cf3031286 fix the propagation of color space info in decoder
This addresses the issue #960

Change-Id: Iddf45b4bd4f53cb0ddfd879e800a071cd843b915
2015-02-23 13:01:14 -08:00
Hangyu Kuang
bdd249be31 Optimize the dequantization process on decoder side.
Change-Id: I00621ff7165bbe86a18794b4a816976c9effaf78
2015-02-19 15:43:15 -08:00
hkuang
bf3cb25019 Remove unnecessary border extension when frame size change.
This border extension is not needed with on-demond border extension.

Change-Id: I8501b37f5f756dc7e874cef4c1cfdbfa9a16112a
2015-02-10 14:55:27 -08:00
Yunqing Wang
4ae092c660 Make encoder and decoder share common thread function
Moved vp9_accumulate_frame_counts to vp9_thread_common.c to
eliminate the duplicate code.

Change-Id: I9cf506d729603c8bf1494b4c86a3b7d47af1917a
2015-02-06 11:45:51 -08:00
Yunqing Wang
41063137c3 Rename loopfilter_thread files to thread_common files
Renames the files to allow more common thread code to
be moved to vp9/common.

Change-Id: I7386e64e221086e3cdc087e79812f993c423413b
2015-02-06 10:03:31 -08:00
Yunqing Wang
b3b7645a2f vp9_dthread: remove frame_parallel_decoding_mode requirement
This patch continues the work to remove frame_parallel_decoding_mode
requirement in VP9 multi-threaded tile decoder. In order to do that,
the frame counts associated to each thread need to be accumulated
together after the frame is decoded.

Change-Id: Idba1a756cedfed3c154aef52ed82c8da3bbf9e0c
2015-02-04 09:16:41 -08:00
Yunqing Wang
85a9bc04d4 vp9_dthread: pass frame counts to decoder functions
The current multi-threaded tile decoder requires that the videoes
are encoded with frame_parallel_decoding_mode = 1. This requirement
is not necessary, and is better to be removed. This patch includes
the first part of the work.

Change-Id: Ic7695fb3cfe13f9022582c9f0edd2aa6e2e36d28
2015-02-03 09:39:15 -08:00
hkuang
94a459522e Fix a bug from merging frame parallel branch into master.
The merge did not merge the fix for issue #850.

Change-Id: I0dc1377dbfcb9497fb01a13d4f78ac65bff5eb33
2015-02-02 16:01:17 -08:00
hkuang
be6aeadaf4 Try again to merge branch 'frame-parallel' into master branch.
In frame parallel decode, libvpx decoder decodes several frames on all
cpus in parallel fashion. If not being flushed, it will only return frame
when all the cpus are busy. If getting flushed, it will return all the
frames in the decoder. Compare with current serial decode mode in which
libvpx decoder is idle between decode calls, libvpx decoder is busy
between decode calls.

Current frame parallel decode will only speed up the decoding for frame
parallel encoded videos. For non frame parallel encoded videos, frame
parallel decode is slower than serial decode due to lack of loopfilter
worker thread.

There are still some known issues that need to be addressed. For example:
decode frame parallel videos with segmentation enabled is not right sometimes.

* frame-parallel:
  Add error handling for frame parallel decode and unit test for that.
  Fix a bug in frame parallel decode and add a unit test for that.
  Add two test vectors to test frame parallel decode.
  Add key frame seeking to webmdec and webm_video_source.
  Implement frame parallel decode for VP9.
  Increase the thread test range to cover 5, 6, 7, 8 threads.
  Fix a bug in adding frame parallel unit test.
  Add VP9 frame-parallel unit test.
  Manually pick "Make the api behavior conform to api spec." from master branch.
  Move vp9_dec_build_inter_predictors_* to decoder folder.
  Add segmentation map array for current and last frame segmentation.
  Include the right header for VP9 worker thread.
  Move vp9_thread.* to common.
  ctrl_get_reference does not need user_priv.
  Seperate the frame buffers from VP9 encoder/decoder structure.
  Revert "Revert "Revert "Revert 3 patches from Hangyu to get Chrome to build:"""
 Conflicts:
       test/codec_factory.h
       test/decode_test_driver.cc
       test/decode_test_driver.h
       test/invalid_file_test.cc
       test/test-data.sha1
       test/test.mk
       test/test_vectors.cc
       vp8/vp8_dx_iface.c
       vp9/common/vp9_alloccommon.c
       vp9/common/vp9_entropymode.c
       vp9/common/vp9_loopfilter_thread.c
       vp9/common/vp9_loopfilter_thread.h
       vp9/common/vp9_mvref_common.c
       vp9/common/vp9_onyxc_int.h
       vp9/common/vp9_reconinter.c
       vp9/decoder/vp9_decodeframe.c
       vp9/decoder/vp9_decodeframe.h
       vp9/decoder/vp9_decodemv.c
       vp9/decoder/vp9_decoder.c
       vp9/decoder/vp9_decoder.h
       vp9/encoder/vp9_encoder.c
       vp9/encoder/vp9_pickmode.c
       vp9/encoder/vp9_rdopt.c
       vp9/vp9_cx_iface.c
       vp9/vp9_dx_iface.c

This reverts commit a18da9760a.

Change-Id: I361442ffec1586d036ea2e0ee97ce4f077585f02
2015-01-30 21:00:13 -08:00
Johann
a18da9760a Revert "Merge branch 'frame-parallel' to enable frame parallel decode in master branch."
This reverts commit bde04ce503

Change-Id: I053dae04c761b04a36dc239558503905a14d2470
2015-01-23 08:42:02 -08:00
hkuang
bde04ce503 Merge branch 'frame-parallel' to enable frame parallel decode in master branch.
In frame parallel decode, libvpx decoder decodes several frames on all
cpus in parallel fashion. If not being flushed, it will only return frame
when all the cpus are busy. If getting flushed, it will return all the
frames in the decoder. Compare with current serial decode mode in which
libvpx decoder is idle between decode calls, libvpx decoder is busy
between decode calls. VP9 frame parallel decode is >30% faster than serial
decode with tile parallel threading which will makes devices play 1080P
VP9 videos more easily.

* frame-parallel:
  Add error handling for frame parallel decode and unit test for that.
  Fix a bug in frame parallel decode and add a unit test for that.
  Add two test vectors to test frame parallel decode.
  Add key frame seeking to webmdec and webm_video_source.
  Implement frame parallel decode for VP9.
  Increase the thread test range to cover 5, 6, 7, 8 threads.
  Fix a bug in adding frame parallel unit test.
  Add VP9 frame-parallel unit test.
  Manually pick "Make the api behavior conform to api spec." from master branch.
  Move vp9_dec_build_inter_predictors_* to decoder folder.
  Add segmentation map array for current and last frame segmentation.
  Include the right header for VP9 worker thread.
  Move vp9_thread.* to common.
  ctrl_get_reference does not need user_priv.
  Seperate the frame buffers from VP9 encoder/decoder structure.
  Revert "Revert "Revert "Revert 3 patches from Hangyu to get Chrome to build:"""

 Conflicts:
       test/codec_factory.h
       test/decode_test_driver.cc
       test/decode_test_driver.h
       test/invalid_file_test.cc
       test/test-data.sha1
       test/test.mk
       test/test_vectors.cc
       vp8/vp8_dx_iface.c
       vp9/common/vp9_alloccommon.c
       vp9/common/vp9_entropymode.c
       vp9/common/vp9_loopfilter_thread.c
       vp9/common/vp9_loopfilter_thread.h
       vp9/common/vp9_mvref_common.c
       vp9/common/vp9_onyxc_int.h
       vp9/common/vp9_reconinter.c
       vp9/decoder/vp9_decodeframe.c
       vp9/decoder/vp9_decodeframe.h
       vp9/decoder/vp9_decodemv.c
       vp9/decoder/vp9_decoder.c
       vp9/decoder/vp9_decoder.h
       vp9/encoder/vp9_encoder.c
       vp9/encoder/vp9_pickmode.c
       vp9/encoder/vp9_rdopt.c
       vp9/vp9_cx_iface.c
       vp9/vp9_dx_iface.c

Change-Id: Ib92eb35851c172d0624970e312ed515054e5ca64
2015-01-22 18:18:53 -08:00
Yunqing Wang
e76eaf05b1 vp9_ethread: add parallel loopfilter
1. Added row-based loopfilter in encoder;
2. Moved common multi-threaded loopfilter functions from decoder
   to common;
3. Merged multi-threaded loopfilter code, and made encoder/
   decoder call same function to reduce code duplication.

Encoder tests showed that 1% - 2% speedup was seen for good-quality
2-pass mode(at speed 3); 1% - 3% speedup using 2 threads and 4% - 6%
speedup using 4 threads were seen for real-time mode(at speed 7).

Change-Id: I8a4ac51c2ad9bab9fa7b864e90743931c53ec1c4
2015-01-16 17:19:27 -08:00
Yaowu Xu
e94b415c34 Add encoder control for setting color space
This commit adds encoder side control for vp9 to set color space info
in the output compressed bitstream.

It also amends the "vp9_encoder_params_get_to_decoder" test to verify
the correct color space information is passed from the encoder end to
decoder end.

Change-Id: Ibf5fba2edcb2a8dc37557f6fae5c7816efa52650
2015-01-14 10:17:14 -08:00
Yaowu Xu
6b223fcb58 Enable decoder to pass through color space info
This commit added a field to vpx_image_t for indicating color space,
the field is also added to YUV_BUFFER_CONFIG. This allows the color
space information pass through the decoder from input stream to the
output buffer.

The commit also updated compare_img() function with added verification
of matching color space to ensure the color space information to be
correctly passed from encode to decoder in compressed vp9 streams.

Change-Id: I412776ec83defd8a09d76759aeb057b8fa690371
2015-01-13 15:13:19 -08:00